Abstract
Background: The purpose of this study was to evaluate the diagnostic accuracy of clinical tests and pre-operative MRI in detecting subscapularis (SSC) tendon tears identified during arthroscopy.
Methods: Eighty-eight consecutive patients (90 shoulders) who underwent rotator cuff repair were evaluated retrospectively. SSC tear size was classified according to the classification system of Fox. Diagnostic values (sensitivity, specificity, accuracy, positive and negative likelihood ratio) of lift-off test (LOT), belly-press test (BPT), manual muscle strength of internal rotation (MMT), and MRI were calculated. Those of combined clinical tests were also calculated. In addition, sensitivity of each clinical test was compared between the smaller tear group (Fox type I, II) and the larger tear group (type III, IV).
Results: SSC tendon tears were identified in 31 shoulders (34%) during shoulder arthroscopy. All the clinical tests and MRI showed relatively low sensitivity and high specificity. Although positive likelihood ratio of combined BPT and LOT (7.0) was higher than others, we could not recognize any combinations of these tests which brought about an increase of diagnostic performance. Sensitivity of clinical tests and MRI were high enough for detecting large tears, but those for the small tears were less than 50%.
Conclusion: All the clinical tests and MRI showed low sensitivity and high specificity. Diagnostic performance was not increased by the combination of these tests.
